STM32F417VGT VCAP Voltage Problem

[SalihKun]

Hi everyone!

I am having a hard time making my design work.

I prepared a long list of what I did and what I observed to paste into ChatGPT.
I am pasting the exact text below:

I designed an digital circuit using STM32F417VGT microcontroller.

I applied 3.3V to all VDD, VDDA, VBAT and VREF+ pins.
I connected all VSS and VSSA pins to the circuit ground (GND).
I connected all required bypass capacitors to the power pins as instructed in the datasheet.
I added 2.2uF/6.3V ceramic capacitors to the both VCAP pins.
I added 32.768kH and 25MHz crystal oscillators to the correct clock pins.
I connected the BOOT1 pin to GND.
I tried connecting the BOOT0 pin to VDD and GND at separate times.

I have several similar designs, which I previously designed before this one.
In these similar designs, I exactly did the same connections I listed above.
These similar designs work perfectly fine.

But my current design does not work.

I am using StmCubeMxIde (latest version).

I checked the pin connections several times.
I can confirm that there is no mistake in the hardware design.
The supply voltage (3.3V) is very close the actual 3.3V value and is inside the tolerance.

But there is an unexpected behavior I detected:
VCAP pin voltages are well below the value they should be.
VCAP capacitors must have around 1.2V on them.
But they have a voltage value around 200mV-400mV on them.

I checked if the VCAP capacitors are broken.
I read several kilo ohms of resistance between the capacitor terminals.

I am using the standard 20 pin JTAG/SWD connector to program the MCU.
All of the signals below are correctly connected to the connector:
SYS_JTRST
SYS_JTDO-SWO
SYS_JTDI
SYS_JTCK-SWCLK
SYS_JTMS-SWDIO
NRST
The programming interface is the same as my previously running designs.

In the IDE;
I created a new "STM32 Project".
I selected my chip model (STM32F417VGT).
I chose target language C++.
I didn't change any other settings while creating a new project.

In the "Pins & Configuration" tab:
I assigned the pins for crystal connections. (System Core > RCC > HSE/LSE)
I assigned the pins for JTAG/SWD connections. (System Core > SYS > Debug > JTAG (5 pins))

In the "Pins & Configuration" tab:
Input frequencies (32.768kHz and 25MHz) are correctly listed.
There are no errors reported.

The IDE generated a default code for me.
I didn't do any modifications in this default code.

When I click either "Debug" or "Run" button in the toolbar, I get the following error message in the console panel:
"Target no device found"
"Error in initializing ST-LINK device."
"Reason: No device found on target."

In the Debug/Run configuration page:
I can't scan the programmer STLink S/N.
"Debuge probe" --> "STL-LINK (ST LING GDB server)"
"Interface" --> "SWD"
"Frequency" --> "Auto"
"Reset behavior > Type" --> "Connect under reset"

After a short while, a message box with the title "Problem Occurred" appears with these messages:
"'Launching <ProjectName> Debug' has encountered a problem."
"Error in final launch sequence:"
"Failed to execute MI command:
target remote localhost:61234"
"Error message from debugger back end:
localhost:61234: Connection timed out."

My programmer is "ST-LINK/V2".
Its firmware is up to date.
Its led color is always red.

When I use "STM32CubeProgrammer" software:
It can't connect to the target.
It measures target voltage as "3.25V".
It reads target version as "V2J45S7".

I have assembled two PCBs of this same design.
Both of them are showing the exact same symptoms.

Can you please guide me what to do in order to solve this issue?

Thanks.

[liaifat85]

There should be  2.2 µF, 6.3V ceramic capacitors, and placed as close as possible to the VCAP pins. 

[SalihKun]

I exactly did so.
I connected one 2.2µF, 6.3V ceramic capacitor to each VCAP pin.

[STOne-32]

Dear @SalihKun ,

Share you schematics , please . And also a picture of the device Top marking to be sure it is genuine .

Ciao

STOne-32

[Tesla DeLorean]

Check Part Orientation

>>I can confirm that there is no mistake in the hardware design.

I'd suppose if that were true it would be working now.

Hard to make an independent assessment without a schematic, and implementation details.

Could be fake chips, you got these through a reputable distributor?

[gbm]

Verify that ALL pins with names starting with Vdd ale connected to 3.3V and ALL GND pins are connected to the ground. Do it first with the schematic then with the board, using ohmmeter's continuity check function.

[SalihKun]

Hello;

I am sorry for the late reply.
I am attaching schematics and an image of the chip recorded during assembly.

Looking for your help.
Thanks.

EDIT: Oh god, now that I am looking at the attached image "Chip under microscope.png", don't tell me I messed up with the chip orientation? Can anyone confirm it for me? Please!

[gbm]

Not only it's rotated by 90 degrees but also there are shorts between top left pins. Easy thing - should take no more than 5 minutes with hot air station and good flux.

[SalihKun]

I know that the smaller dot which I marked as "A". But I want to make sure of it. Can you please confirm it by telling me which one is the number 1 mark?

A or B?

[Tesla DeLorean · Accepted solution]

A is the Pin 1 designation, so the text is read-able western style. Pin 1 is on the edge Below the Mark

B is the Ejection Mold / Relief, these can be different depending on the production plant / line

Solder Paste be everywhere..